The distribution of substance P and neuropeptide Y in four songbird species: a comparison of food-storing and non-storing birds.

The distributions of the neuropeptides substance P (SP) and neuropeptide Y (NPY) were investigated in four songbird species that differ in their food-storing behavior. The food-storing black-capped chickadee (Parus atricapillus) was compared to the non-storing blue tit (Parus caeruleus) and great tit (Parus major) within the avian family Paridae, as well as to the non-storing dark-eyed junco (Junco hyemalis). All four species showed a similar distribution of SP throughout the brain with the exception of two areas, the hippocampal complex (including hippocampus (Hp) and parahippocampus (APH)) and the Wulst (including the hyperstriatum accessorium (HA)). SP-like immunoreactivity was found in cells of the Hp in juncos, but not in the three parid species. Two areas within the APH and HA showed SP-like immunoreactivity in all four species. The more medial of these (designated SPm) is a distinctive field of fibers and terminals found throughout the APH and extending into the HA. A positive relationship between SPm and Hp volume was found for all four species with the chickadee having a significantly larger SPm area relative to telencephalon than the other species. The distribution of SP in this region may be related to differences in food-storing behavior. In contrast to substance P, NPY distribution throughout the brain was similar in all four species. Further, NPY-immunoreactive cells were found in the Hp of all four species and no species differences in the number of NPY cells was observed.

The medial extended amygdala in male reproductive behavior. A node in the mammalian social behavior network.

Hormonal and chemosensory signals regulate social behaviors in a wide variety of mammals. In the male Syrian hamster, these signals are integrated in nuclei of the medial extended amygdala, where olfactory and vomeronasal system transmission is modulated by populations of androgen- and estrogen-sensitive neurons. Evidence from behavioral changes following lesions and from immediate early gene expression supports the hypothesis that the medial extended amygdala and medial preoptic area belong to a circuit that functions selectively in male sexual behavior. However, accumulated behavioral, neuroanatomical, and neuroendocrine data in hamsters, other rodents, and other mammals indicate that this circuit is embedded in a larger integrated network that controls not only male mating behavior, but female sexual behavior, parental behavior, and various forms of aggression. In this context, perhaps an individual animal's social responses can be more easily understood as a repertoire of closely interrelated, hormone-regulated behaviors, shaped by development and experience and modulated acutely by the environmental signals and the hormonal milieu of the brain.

Androgen receptor immunoreactivity in the male and female Syrian hamster brain.

To investigate potential mechanisms for sex differences in the physiologic response to androgens, the present study compared the hormonal regulation of intracellular androgen receptor partitioning and the distribution of androgen receptor immunoreactivity in select brain regions from male and female hamsters. Androgen receptors were visualized on coronal brain sections. Two weeks after castration, androgen receptor immunoreactivity filled the neuronal nuclei and cytoplasm in males and females. In gonad-intact males and females, androgen receptor immunoreactivity was limited to the cell nucleus. Whereas exogenous dihydrotestosterone prevented cytoplasmic immunoreactivity, estrogen at physiologic levels did not. These results suggest that nuclear androgen receptor immunoreactivity in gonad-intact females is maintained by endogenous androgens, and that androgens have the potential to influence neuronal activity in either sex. However, sex differences in the number and staining intensity of androgen-responsive neurons were apparent in select brain regions. In the ventral premammillary nucleus, ventromedial nucleus of the hypothalamus, and medial amygdaloid nucleus, androgen receptor staining was similar in gonadectomized males and females. In the lateral septum, posteromedial bed nucleus of the stria terminalis (BNSTpm), and medial preoptic nucleus, the number of androgen receptor-immunoreactive neurons was significantly lower in females (p < .05). Moreover, the integrated optical density/cell in BNSTpm was significantly less in females (1.28+/-0.3 units) than in males (2.21+/-0.2 units; p < .05). These sex differences in the number and staining intensity of androgen-responsive neurons may contribute to sex differences in the behavioral and neuroendocrine responses to androgens.

Neural pathway from the olfactory bulbs regulating tonic gonadotropin secretion.

Removal of the olfactory bulbs of male golden hamsters results in a marked increase in tonic gonadotropin, prolactin and testosterone secretion which counteracts inhibitory effects of manipulations such as maintenance on short photoperiod, food restriction or treatment with gonadal steroids. The bulbectomy-induced increase in hormone secretion is interpreted to reflect a tonic inhibitory influence of the olfactory bulbs. This inhibition is not dependent upon chemosensory stimulation and may be mediated by olfactory bulb fibers projecting through the lateral olfactory tract to or through the olfactory tubercle. This review will relate these studies conducted on hamsters to results in other species, such as the rat, where the olfactory bulbs enhance serum gonadotropin levels.

Fos-immunoreactivity within the extended amygdala is correlated with the onset of sexual satiety.

We hypothesized that c-fos expression in the medial amygdala (Me), the bed nucleus of the stria terminalis (BNST), and the medial preoptic area (MPOA) of the male Syrian hamster brain correlated with sexual satiety. To address this hypothesis, males were mated for 4 consecutive days. Experiment 1 determined whether the number of Fos-immunoreactive (Fos-ir) nuclei was equivalent in two groups of males mated to sexual satiety, one group of rested males (9.67 +/- 0.80 ejaculations) and a second group mated for 4 consecutive days (3.50 +/- 0.56 ejaculations). Fos-ir was increased within the caudal posterodorsal Me (cMePD), the anterodorsal and posteroventral subdivisions of the posteromedial BNST [BNSTpm(ad) and BNSTpm(pv)], the dorsolateral MPOA, and the medial preoptic nucleus of all males mated to sexual satiety compared to nonmated controls. In addition, Fos-ir "clusters" within the cMePD and BNSTpm(ad) were present in males mated to satiety regardless of the number of ejaculations. However, all males achieved multiple ejaculations. Therefore, Experiment 2 examined whether two groups of males stopped at one ejaculation exhibit different patterns of Fos-ir depending on proximity to sexual satiety. Brains of consecutively mated males, closer to satiety than rested males, showed greater BNSTpm(pv) Fos-ir and 5/6 males, but no rested male, exhibited cMePD Fos-ir clusters. These results support the hypothesis that cMePD and BNSTpm(pv) neuronal activation is associated with satiety and may constitute a discrete circuit to terminate mating.

Non invasive in vivo anatomical studies of the oscine brain by high resolution MRI microscopy.

We describe in this paper an in vivo Magnetic Resonance Imaging (MRI) procedure that allows one to obtain three-dimensional high quality images of the entire brain of small passerine birds such as the canary with a slice thickness of 58 micron and an image resolution of 78 microns. This imaging procedure was completed in 70 min on anaesthetised birds that later recovered uneventfully and could be reused for subsequent additional imaging. To illustrate the high resolution and anatomical detail that can be achieved, examples of coronal images through the entire hypothalamus are provided in the same sectioning plane as the previously published canary brain atlas. The data set can be used to create sections in any desired plane and the entire data set can be viewed from any point of view in a volume rendered image. This provides a useful tool in understanding the three-dimensional organisation of the brain. Similar procedures can also be applied on fixed brains and might allow an even better anatomical resolution of images because time constrains no longer limit the duration of image acquisition. The non-invasive MRI technique enables to study neuroanatomical features with a high resolution and without killing the animal subjects so that measures can be obtained in a same individual both before and after an experimental treatment.

Mating-induced expression of c-fos in the male Syrian hamster brain: role of experience, pheromones, and ejaculations.

This study was designed to investigate the effects of pheromonal cues and specific behaviors within the male copulatory sequence on c-fos expression in the medial nucleus of the amygdala (Me), the bed nucleus of the stria terminalis (BNST), and the medial preoptic area (MPOA) of the Syrian hamster brain. Sexually experienced male hamsters were placed into clean testing arenas and were either: 1) left alone as handled controls; 2) exposed to female hamster vaginal secretion (FHVS) on cotton swabs; or mated to various end points of copulation with a sexually receptive female: 3) five intromissions, 4) one ejaculation, 5) five ejaculations, or 6) long intromissions, A seventh group of sexually naive control males 7) was left alone in the arena. The brains of these males were compared to those of the sexually experienced controls to determine whether exposure to cues associated with prior sexual experience could alter c-fos expression. In males exposed only to FHVS, Fos immunoreactivity (Fos-ir) increased within the posterodorsal Me, the anterodorsal part of the posteromedial BNST, and the magnocellular medial preoptic nucleus (MPNmag). Following one ejaculation, Fos-ir increased within the caudal posterodorsal Me, the dorsolateral MPOA, and the paraventricular nucleus of the hypothalamus. After multiple ejaculations, additional labeling was observed within the posteroventral part of the posteromedial BNST, the medial preoptic nucleus (MPN), the central tegmental field, and in cell clusters of the caudal posterodorsal Me and rostral posteromedial BNST. Fos-ir also increased within the posterodorsal Me, MPN, and MPNmag in sexually experienced control males exposed to the empty test chamber compared to sexually naive males exposed to an identical chamber. These results demonstrate that the mating-induced pattern of neuronal activation in sexually experienced males is dependent upon multiple factors, including prior sexual experience in the testing environment, investigation of FHVS, and the number of ejaculations achieved.

Lesions of the ventral striatum mimic the effect of olfactory bulbectomy to prevent short photoperiod-induced testicular regression in golden hamsters.

Bilateral olfactory bulbectomy (BX) or bilateral transection of the rostral lateral olfactory tract (LOT) at the level of the anterior olfactory nucleus markedly increases gonadotropin secretion and prevents the testicular regression associated with maintenance on short photoperiod in golden hamsters. In an effort to further elucidate the neural tracts involved in this influence on gonadotropin secretion, lesions were placed in several potential pathways. Hamsters underwent sham surgery (SH), bilateral BX, or electrolytic or radiofrequency lesions of the: medial nucleus of the amygdala (MeX) caudal LOT just rostral to the medial nucleus of the amygdala (LOTX); or ventral striatum (VSX). Lesions were either bilateral or unilateral with contralateral olfactory bulbectomy. All animals were then placed on short photoperiod (LD 10:14) for 10 weeks and testicular size and body weight were assessed at weekly intervals. Lesion placement was assessed in brain sections stained with cresyl violet and animals with misplaced lesions were excluded. The following represent the number of animals in each group undergoing testicular regression in response to short photoperiod: SH: 32/35; BX: 8/31 (P < 0.01 vs. SH); MeX: 5/5; caudal LOTX: 8/9 and VSX: 3/8 (P < 0.05 vs. SH). Serum LH, FSH and testosterone at the end of the study correlated with the testicular regression data. These results suggest that the tonic inhibitory effect of the olfactory bulbs on gonadotropin secretion is mediated by fibers that exit the LOT rostral to the amygdala and project medially, either passing through or synapsing in the ventral striatum.

Androgen and estrogen receptors coexist within individual neurons in the brain of the Syrian hamster.

Many aspects of reproductive neuroendocrine function and sexual behavior are responsive to both androgens and estrogens, suggesting that receptors for these steroid hormones may reside within single cells in brain regions that control reproductive function. We determined the distribution of estrogen receptor-containing neurons in 40-microns coronal brain sections in gonadectomized, DHT-treated male Syrian hamsters using immunocytochemistry with the H222 antibody (10 micrograms/ml; Abbott Laboratories). Subsequently, we colocalized estrogen receptors with androgen receptors using the PG-21 antibody (0.5 microgram/ml; G.S. Prins). In males, the distribution of estrogen receptor-containing neurons was similar to that reported previously for the female hamster. Colocalization of androgen and estrogen receptor immunoreactivity was observed in brain regions that contain numerous androgen and estrogen receptor-positive neurons, including subdivisions of the medial preoptic area, bed nucleus of the stria terminals, ventromedial nucleus of the hypothalamus, and the amygdalohippocampal area. Single-labelled estrogen receptor-containing neurons were most numerous in the amygdalohippocampal area and the rostral medial preoptic nucleus; androgen receptor-immunoreactive cells were most abundant in the ventral premammillary nucleus and the lateral septum. These data suggest the potential of androgens and estrogens to influence neuronal function within individual steroid receptor-containing neurons of the hamster limbic system.

Integration of chemosensory and hormonal cues is essential for mating in the male Syrian hamster.

Mating behavior in the male hamster is dependent upon both chemosensory and hormonal cues, and copulation is abolished if either signal is interrupted. Through reciprocal interactions of these signals, chemosensory stimuli increase circulating testosterone in the male, and the male's hormonal status influences his attraction to female pheromones. Furthermore, anatomical data suggest that these signals are transmitted through parallel pathways in separate subdivisions of the medial amygdaloid nucleus, bed nucleus of the stria terminalis (BNST) and medial preoptic area (MPOA). The aim of this study was to determine if the integration of chemosensory and hormonal cues is essential for mating. We combined an intracerebral implant of testosterone in BNST/MPOA with removal of a single olfactory bulb (UBx), ipsilateral or contralateral to the steroid implant. Previous studies have demonstrated that testosterone implants which stimulate androgen receptor-containing neurons in posteromedial BNST and MPOA can increase mounting in males castrated for 12 weeks. Moreover, unilateral bulbectomy alone does not prevent mating. In the present study, ipsilateral UBx prevents communication between hormonal and chemosensory circuits. Sexually experienced males were used. Twelve weeks after castration, a single olfactory bulb was removed, and each male received a testosterone-filled cannula (23 ga) directed at the MPOA. Two weeks later, sexual activity increased in six males with implants in BNST/MPOA and contralateral UBx, but copulation was not restored in eight males with ipsilateral UBx despite equivalent implant placement. This study demonstrates that communication between neurons receiving hormonal signals and chemosensory cues is required for mating behavior.

The medial amygdaloid nucleus and medial preoptic area mediate steroidal control of sexual behavior in the male Syrian hamster.

An important goal of studies on steroid receptors in the brain is to understand the functions of specific populations of steroid receptor-containing neurons, particularly in the control of sexual behavior. The present study compared the ability of testosterone implants directed toward the medial amygdaloid nucleus (Me) or the bed nucleus of the stria terminalis (BNST) and medial preoptic area (MPOA) to stimulate mating in castrated males. Twenty adult male hamsters were selected for vigorous sexual behavior, and baseline measures of copulatory behavior during the first 10 min of a 30-min mating behavior test were recorded on two occasions. Twelve weeks after castration, sexual behavior was recorded as before, and each male received a single intracranial implant constructed of 23-gauge tubing packed with crystalline testosterone placed stereotaxically into Me or BNST/MPOA (n = 10 each). Behavior was assessed on two occasions after surgery to determine if implants in Me or BNST/MPOA could stimulate sexual behavior above that observed after castration. In half of the males from each group, testosterone increased the males' interaction with the female, anogenital investigation, mounting, and reduced the latency to the first mount. After completion of behavioral testing, males were perfused and the brains processed for androgen receptor immunocytochemistry to determine the extent of brain regions influenced by the testosterone-filled cannula. In tissues stained rapidly in diaminobenzidine, androgen receptor-containing neurons were visible only in steroid responsive brain regions adjacent to the testosterone implant. This approach revealed that steroid receptors in both the posterior subdivision of Me and in the medial subregions of BNST/MPOA can mediate hormonal control of mating behavior in the male Syrian hamster. Implants placed outside these regions failed to stimulate mating. These results suggest that maintenance of copulatory behavior by gonadal steroids is not transduced by a single brain region, but that steroid effects can be elicited at multiple points along the mating behavior pathway.

Mating and agonistic behavior produce different patterns of Fos immunolabeling in the male Syrian hamster brain.

Previous work has shown that mating induces the expression of Fos protein within the chemosensory pathways of the male Syrian hamster brain. However, it is not known if this pattern of labeling is specific to mating or the result of social interactions in general. To determine the behavioral specificity of activation within these pathways, Fos immunostaining following mating was compared to that following agonistic behavior. Both mating and agonistic behavior are dependent upon chemosensory cues and gonadal steroids (reviewed in Refs 64, 65) and areas belonging to the olfactory and vomeronasal pathways process chemosensory and hormonal information (reviewed in Ref. 48). The results of this study demonstrate both similarities and differences in brain activation patterns following these two social behaviors. Agonistic behavior increased the number of Fos-immunoreactive neurons within most subdivisions of the medial amygdala, the anteromedial and posterointermediate bed nucleus of the stria terminalis, the ventrolateral septum and the ventral premammillary nucleus of the hypothalamus in a pattern comparable to that observed after mating. This pattern of activation common to mating and agonistic behavior may reflect an increase in an animal's general state of arousal during social interactions. In contrast, although mating and agonistic behavior both activated neurons within the caudal subdivision of the medial nucleus of the amygdala, the anterodorsal level of posteromedial bed nucleus of the stria terminalis and the paraventricular and ventromedial nuclei of the hypothalamus, in these areas either the distribution and/or number of Fos-immunoreactive neurons differed. In addition, agonistic behavior selectively activated neurons within the anterolateral bed nucleus of the stria terminalis, the anterior nucleus of the hypothalamus and the dorsal periaqueductal gray, whereas mating alone activated neurons within the posteroventral level of posteromedial bed nucleus of the stria terminalis and the medial preoptic area. No differences were found between dominant and subordinate males following agonistic behavior. These observations along with results from other laboratories suggest that mating and agonistic behavior activate distinct neural circuits.

Axon-sparing lesions of the medial nucleus of the amygdala decrease affiliative behaviors in the prairie vole (Microtus ochrogaster): behavioral and anatomical specificity.

The neural basis of affiliative behavior was examined in the prairie vole, a rodent that exhibits high levels of social contact and paternal behavior. In the first study, the axon-sparing excitotoxin N-methyl-D,L-aspartic acid (NMA) produced lesions in the basolateral nucleus of the amygdala or the corticomedial amygdala. Males with corticomedial lesions showed significantly less contact with a familiar adult female and a pup when compared with males with lesions of the basolateral nucleus or controls. This behavioral change was not associated with changes in exploratory behavior, motor function, performance in an olfactory task, fearfulness, physical well-being, or body temperature. In a second study, NMA lesions restricted to the medial nucleus also decreased paternal behavior. Neurons in the medial nucleus of the amygdala appear to be essential for the normal expression of paternal care in this species.

Colocalization of tyrosine hydroxylase and Fos in the male Syrian hamster brain following different states of arousal.

In an investigation of the role that central tyrosine hydroxylase-(TH) containing neurons play in copulation in the male Syrian hamster, the induction of Fos protein was used as an index of neuronal activation. With a double immunoperoxidase technique, the activation of TH neurons was compared in hamsters from three experimental groups: (1) mated in a new cage; (2) handled controls placed into a new cage, and (3) unhandled controls. Although mating selectively induces Fos production in the medial amygdaloid nucleus (Me), more than half of the TH neurons in Me (a region outside of the classical catecholamine systems) expressed Fos equally in all of the experimental groups. In the paraventricular hypothalamic nucleus (PVN), TH neurons were activated equivalently in mated and handled control animals compared to unhandled controls. TH neurons in the nucleus of the solitary tract (NST) were also activated in handled control animals, and mating further enhanced the level of Fos immunostaining in these neurons above both groups of nonmated animals. Although not quantified, co-localization of Fos and TH was also observed in all experimental groups in the olfactory bulbs and the interfascicular nucleus, and in the horizontal limb of the diagonal band of Broca and the cerebral cortex, regions which contain TH neurons but are not part of the classically described TH cell groups. Few, if any, TH neurons in other catecholaminergic brain regions, such as the substantia nigra and locus coeruleus, produced Fos in any of the experimental groups. These results suggest that TH neurons in the PVN and NST may be activated during different states of arousal, and that nonclassical TH neurons in the amygdala produce high levels of Fos even in unstimulated animals.

Tyrosine hydroxylase mRNA-containing neurons in the medial amygdaloid nucleus and the reticular nucleus of the thalamus in the Syrian hamster.

To confirm previous immunocytochemical findings in colchicine-treated Syrian hamsters, in situ hybridization was used to investigate the distribution of TH mRNA-containing cells in the medial amygdaloid nucleus (Me) and the thalamic reticular nucleus (Rt) of untreated hamsters. TH mRNA-producing neurons were observed in anterior and posterior Me and throughout Rt, similar to the distribution of TH-immunostained cells in these areas of animals receiving colchicine. These data confirm that TH is normally produced in amygdaloid and thalamic cell groups which lie outside the classical catecholamine systems.

Intracellular partitioning of androgen receptor immunoreactivity in the brain of the male Syrian hamster: effects of castration and steroid replacement.

The effect of castration and steroid replacement on the intracellular partitioning of the androgen receptor in the brain of the male Syrian hamster was determined using immunocytochemistry. Androgen receptors were visualized using the PG-21 antibody (G. S. Prins) on 40-microns coronal brain sections from hamsters perfused with 4% paraformaldehyde with or without 0.4% glutaraldehyde. Control studies confirmed antibody specificity in gonad-intact and castrate males. In the normal adult male, androgen receptor immunocytochemistry reveals intense staining confined to the cell nucleus. Castration caused a gradual increase in cytoplasmic labelling within 2 weeks, accompanied by a reduction in nuclear staining intensity in androgen receptor-containing neurons throughout the brain. Cytoplasmic androgen receptor staining was eliminated after treatment of orchidectomized males for only 8 h with exogenous testosterone. Likewise, long-term exposure to testosterone and dihydrotestosterone, a nonaromatizable androgen, maintained nuclear androgen receptor immunoreactivity. However, exposure to low physiologic concentrations of estrogen was not effective in this regard. In addition, we determined that nuclear androgen receptor immunoreactivity decreases in response to inhibitory short-day photoperiod, but without an increase in cytoplasmic immunostaining. This appears to be due to the decrease in androgen production by the testis, rather than a direct photoperiodic effect, because testosterone supplementation to short-day males restored the intensity of nuclear androgen receptor immunoreactivity to levels comparable to those in the intact male. These findings are compatible with a new model for the intracellular localization of androgen receptors, in which a subset of unoccupied receptors is located in the cell cytoplasm in the absence of ligand. They further demonstrate the repartitioning of such cytoplasmic receptors, thereby confirming and extending previous observations using biochemical techniques on the regulation of neuronal androgen receptors.

Mating activates androgen receptor-containing neurons in chemosensory pathways of the male Syrian hamster brain.

Fos-immunoreactivity is induced during mating in the male Syrian hamster in limbic areas that relay chemosensory information and contain receptors for gonadal steroid hormones. The induction of Fos is an index of neuronal activation. After mating, c-fos expression is greatest in subnuclei of the medial amygdaloid nucleus (Me), bed nucleus of the stria terminalis (BNST), and medial preoptic area (MPOA). The present study determined if individual neurons in these activated subnuclei contain androgen receptors. We aim to understand how essential chemosensory and hormonal signals are integrated to control copulation. Adult male hamsters (n = 6) were allowed to mate with a sexually receptive female for 30 min. They were perfused 1 h later with 4% paraformaldehyde and 40 microns frozen sections were processed for immunocytochemistry using antisera against Fos (Cambridge Research Biochemicals) and the androgen receptor (G.S. Prins). The brains of three non-mated males were also processed for Fos immunocytochemistry. Mating significantly increased the number of Fos-immunoreactive neurons within subnuclei of Me, BNST, and MPOA relative to non-mated males (P < 0.05). These nuclei contained abundant androgen receptors. In the corticomedial amygdala, 20-40% of Fos-immunoreactive neurons in mated hamsters expressed androgen receptors. Although few androgen receptors are found in the anteromedial and postero-intermediate subdivisions of the BNST, these areas exhibited 26% and 47% co-localization, respectively. In posteromedial BNST, which contains large numbers of steroid receptor-containing neurons, androgen receptors were identified in 48% of Fos-immunoreactive neurons. In the MPOA, 54% of Fos-immunoreactive neurons expressed the androgen receptor throughout the rostrocaudal extent of the medial preoptic nucleus (MPN).(ABSTRACT TRUNCATED AT 250 WORDS)

Tyrosine hydroxylase neurons in the male hamster chemosensory pathway contain androgen receptors and are influenced by gonadal hormones.

Chemosensory and hormonal signals, both of which are essential for mating in the male Syrian hamster, are relayed through a distinct forebrain circuit. Immunocytochemistry for tyrosine hydroxylase, a catecholamine biosynthetic enzyme, previously revealed immunoreactive neurons in the anterior and posterior medial amygdaloid nucleus, one of the nuclei within this pathway. In addition, dopamine-immunoreactive neurons were located in the posterior, but not the anterior, medial amygdala. In the present study, tyrosine hydroxylase-immunostained neurons were also observed in other areas of the chemosensory pathway, including the posteromedial bed nucleus of the stria terminalis and the posterior, lateral part of the medial preoptic area, while dopamine immunostaining was only seen in the posteromedial bed nucleus of the stria terminalis. The colocalization of tyrosine hydroxylase and androgen receptors was examined in these four tyrosine hydroxylase cell groups by a double immunoperoxidase technique. The percentage of tyrosine hydroxylase-immunolabeled neurons that were also androgen receptor-immunoreactive was highest in the posterior medial amygdaloid nucleus (74%) and the bed nucleus of the stria terminalis (79%). Fewer tyrosine hydroxylase-immunostained neurons in the anterior medial amygdala (33%) and the medial preoptic area (4%) contained androgen receptors. Surprisingly, castration resulted in a significant decrease in the number of tyrosine hydroxylase-immunoreactive neurons only in the anterior medial amygdaloid nucleus, and this effect was transient. Six weeks after castration, the anterior medial amygdala contained 61% fewer tyrosine hydroxylase-immunolabeled neurons, but 12 weeks after gonadectomy, immunostaining returned to intact values. The number of immunostained neurons in testosterone-replaced, castrated hamsters was not significantly different from that of intact or castrated animals at any time. The results of this study indicate that a substantial number of tyrosine hydroxylase-immunostained neurons in the chemosensory pathway are influenced by androgens; the majority of these neurons in the posterior medial amygdala and the posteromedial bed nucleus of the stria terminalis produce androgen receptors, and tyrosine hydroxylase immunoreactivity is altered by castration in the anterior medial amygdala.

6-Hydroxydopamine lesions of the nigrostriatal pathway alter the expression of glutamate decarboxylase messenger RNA in rat globus pallidus projection neurons.

In situ hybridization was used to study the effect of 6-hydroxydopamine-induced damage to the midbrain dopaminergic neurons on the level of glutamate decarboxylase mRNA in globus pallidus neurons in the rat. Some animals received an injection of Fluoro-gold in the entopeduncular nucleus or the substantia nigra prior to the 6-hydroxydopamine lesion in order to identify glutamic acid decarboxylase mRNA levels in pallidal neurons that project to one of these targets. Analysis was carried out on a sample of all pallidal neurons as well as neurons that were identified as projection neurons in control and lesioned groups. The loss of the dopamine-containing neurons in the substantia nigra resulted in significant increases in the percentage of globus pallidus neurons that expressed glutamate decarboxylase mRNA and in the amount of glutamate decarboxylase mRNA per globus pallidus neuron. These increases were noted in a sample of all pallidal neurons, as well as pallidal neurons that were identified as projecting to either the entopeduncular nucleus or the substantia nigra. In control animals, glutamate decarboxylase mRNA was clearly identified in globus pallidus neurons projecting to the entopeduncular nucleus, indicating that this recently reported projection is at least partially GABAergic. The results of this study indicate that substantia nigra dopaminergic neurons regulate globus pallidus neurons in the rat, and that removal of the dopaminergic input to the corpus striatum results in a significant increase in the amount of glutamate decarboxylase mRNA in pallidal neurons. The decreased firing rate of pallidal neurons that is seen following the loss of dopamine input appears to be accompanied by an increase in the level of glutamate decarboxylase mRNA in these neurons.